All of lore.kernel.org
 help / color / mirror / Atom feed
From: Ben Myers <bpm@sgi.com>
To: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>,
	Oleg Nesterov <oleg@redhat.com>,
	Linux Kernel <linux-kernel@vger.kernel.org>,
	Alexander Viro <viro@zeniv.linux.org.uk>,
	Dave Jones <davej@redhat.com>,
	xfs@oss.sgi.com
Subject: Re: splice vs execve lockdep trace.
Date: Wed, 17 Jul 2013 18:40:49 -0500	[thread overview]
Message-ID: <20130717234049.GC3572@sgi.com> (raw)
In-Reply-To: <CA+55aFxdqzMY5VJoYaLmL=+=f2s1cbHHV-TjC3=taXpF-xov1w@mail.gmail.com>

Linus,

On Wed, Jul 17, 2013 at 09:03:11AM -0700, Linus Torvalds wrote:
> On Tue, Jul 16, 2013 at 10:51 PM, Dave Chinner <david@fromorbit.com> wrote:
> >
> > But When i say "stale data" I mean that the data being returned
> > might not have originally belonged to the underlying file you are
> > reading.
> 
> We're still talking at cross purposes then.
> 
> How the hell do you handle mmap() and page faulting?

__xfs_get_blocks serializes access to the block map with the i_lock on the
xfs_inode.  This appears to be racy with respect to hole punching.

> Because if you return *that* kind of stale data, than you're horribly
> horribly buggy. And you cannot *possibly* blame
> generic_file_splice_read() on that.

Seems to me we'd need to hold the page lock on every page in the hole to
provide exclusion with splice read and mmap faults, then remove the extents,
and finally truncate the pages away.  I think at that point the reads could be
done without the iolock.  Or, is there a different lock that could do the trick?

Thanks,
	Ben

_______________________________________________
xfs mailing list
xfs@oss.sgi.com
http://oss.sgi.com/mailman/listinfo/xfs

WARNING: multiple messages have this Message-ID (diff)
From: Ben Myers <bpm@sgi.com>
To: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Dave Chinner <david@fromorbit.com>,
	Peter Zijlstra <peterz@infradead.org>,
	Oleg Nesterov <oleg@redhat.com>,
	Linux Kernel <linux-kernel@vger.kernel.org>,
	Alexander Viro <viro@zeniv.linux.org.uk>,
	Dave Jones <davej@redhat.com>,
	xfs@oss.sgi.com
Subject: Re: splice vs execve lockdep trace.
Date: Wed, 17 Jul 2013 18:40:49 -0500	[thread overview]
Message-ID: <20130717234049.GC3572@sgi.com> (raw)
In-Reply-To: <CA+55aFxdqzMY5VJoYaLmL=+=f2s1cbHHV-TjC3=taXpF-xov1w@mail.gmail.com>

Linus,

On Wed, Jul 17, 2013 at 09:03:11AM -0700, Linus Torvalds wrote:
> On Tue, Jul 16, 2013 at 10:51 PM, Dave Chinner <david@fromorbit.com> wrote:
> >
> > But When i say "stale data" I mean that the data being returned
> > might not have originally belonged to the underlying file you are
> > reading.
> 
> We're still talking at cross purposes then.
> 
> How the hell do you handle mmap() and page faulting?

__xfs_get_blocks serializes access to the block map with the i_lock on the
xfs_inode.  This appears to be racy with respect to hole punching.

> Because if you return *that* kind of stale data, than you're horribly
> horribly buggy. And you cannot *possibly* blame
> generic_file_splice_read() on that.

Seems to me we'd need to hold the page lock on every page in the hole to
provide exclusion with splice read and mmap faults, then remove the extents,
and finally truncate the pages away.  I think at that point the reads could be
done without the iolock.  Or, is there a different lock that could do the trick?

Thanks,
	Ben

  reply	other threads:[~2013-07-17 23:40 UTC|newest]

Thread overview: 51+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-07-16  1:53 splice vs execve lockdep trace Dave Jones
2013-07-16  2:32 ` Linus Torvalds
2013-07-16  2:38   ` Dave Jones
2013-07-16  3:25     ` Linus Torvalds
2013-07-16  3:25       ` Linus Torvalds
2013-07-16  3:28       ` Dave Jones
2013-07-16  3:28         ` Dave Jones
2013-07-16  5:31       ` Al Viro
2013-07-16  5:31         ` Al Viro
2013-07-16  6:03       ` Dave Chinner
2013-07-16  6:03         ` Dave Chinner
2013-07-16  6:16         ` Al Viro
2013-07-16  6:16           ` Al Viro
2013-07-16  6:41           ` Dave Chinner
2013-07-16  6:41             ` Dave Chinner
2013-07-16  6:50           ` Dave Chinner
2013-07-16  6:50             ` Dave Chinner
2013-07-16 19:33         ` Ben Myers
2013-07-16 19:33           ` Ben Myers
2013-07-16 20:18           ` Linus Torvalds
2013-07-16 20:18             ` Linus Torvalds
2013-07-16 20:43             ` Dave Chinner
2013-07-16 20:43               ` Dave Chinner
2013-07-16 21:02               ` Linus Torvalds
2013-07-16 21:02                 ` Linus Torvalds
2013-07-17  4:06                 ` Dave Chinner
2013-07-17  4:06                   ` Dave Chinner
2013-07-17  4:54                   ` Linus Torvalds
2013-07-17  4:54                     ` Linus Torvalds
2013-07-17  5:51                     ` Dave Chinner
2013-07-17  5:51                       ` Dave Chinner
2013-07-17 16:03                       ` Linus Torvalds
2013-07-17 16:03                         ` Linus Torvalds
2013-07-17 23:40                         ` Ben Myers [this message]
2013-07-17 23:40                           ` Ben Myers
2013-07-18  0:17                           ` Linus Torvalds
2013-07-18  0:17                             ` Linus Torvalds
2013-07-18  3:42                             ` Dave Chinner
2013-07-18  3:42                               ` Dave Chinner
2013-07-18 21:16                               ` Ben Myers
2013-07-18 21:16                                 ` Ben Myers
2013-07-18 22:21                                 ` Ben Myers
2013-07-18 22:21                                   ` Ben Myers
2013-07-18 22:49                                   ` Dave Chinner
2013-07-18 22:49                                     ` Dave Chinner
2013-07-18  3:17                         ` Dave Chinner
2013-07-18  3:17                           ` Dave Chinner
2013-07-16 13:59       ` Vince Weaver
2013-07-16 13:59         ` Vince Weaver
2013-07-16 15:02         ` Dave Jones
2013-07-16 15:02           ` Dave Jones

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20130717234049.GC3572@sgi.com \
    --to=bpm@sgi.com \
    --cc=davej@redhat.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=oleg@redhat.com \
    --cc=peterz@infradead.org \
    --cc=torvalds@linux-foundation.org \
    --cc=viro@zeniv.linux.org.uk \
    --cc=xfs@oss.sgi.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body. 
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.